There is a chapter on the Rorschach Test in one of William Poundstone’s books. He describes how the test is used to diagnosis everything from schizophrenia to sexual orientation. How accurately it does this is a seperate question.
The ink blots were shown to a great many people who already had an estblished diagnosis. Then the results were examined to see if there were statistically significant instances of people with one particular diagnosis making a particular interpretation or noticing a particular detail when other people didn’t. The test works on the assumption that if one tends to make the interpretations which are “popular” with members of a particular group, the chances are greater that you share their diagnosis.
Poundstone mentions briefly that there has been some important doubts raised about some of the statistical background for the tests. For instance, there is one ink blot which people generally see as two human figures facing each other. According to the traditional interpretation, if either a man or a woman sees them as two men (even though they appear to have large breasts) then one is “probably” homosexual, and if one sees them as two men (even though their genitals appear to be unconnected to the rest of their bodies) one is supposed to be heterosexual. Poundstone notes, however, that a very large survey of people whose sexual orientation was already established showed that a majority of the heterosexual men saw the ink blot as resembling two women.
I haven’t read Poundstone, but this example is nothing like the training I received, or what I teach, or a use I’ve seen the Rorschach put to in at least 20 years within mainstream psychology. “Traditional” here probably means “psychodynamic,” and thogh there are many psychodynamic practitioners, it’s hardly the standard of practice for the field.
In addition to norms for groups with particular diagnoses, it should be noted that the Rorschach has also been normed on large numbers of “normal” people (i.e., no diagnosis).
Responsible testing does not rely on only one instrument. Rather, several different tests are given (a “battery”) and the results examined for correlation. If one instrument gives a very different result, it is incumbent upon the clinician to conduct further inquiry or research, and to interpret and explain this discrepancy meaningfully. I couldn’t ethically diagnose anybody on the grounds of a Rorschach (or MMPI, or BDI, or any single instrument). The most I could do is suggest a diagnostic impression on the basis of that instrument and suggest ways to confirm or disconfirm that impression.
ianzin said: “I’ve done cold reading demos for TV. I’ve had people rate my readings as 99.5% accurate. . .”
I’ve developed a table for rolling D&D character personalities by rolling them a horoscope. When I did this once at a con I got people in line who cooed over how “accurate” it was and how it totally described them even though I had made it clear that I was rolling CHARACTER PERSONALITIES for use in gaming.
I had to do three of these during a round of job applications in Northern Europe. On the last time I followed instructions and asked the (rather delicious) tester if it mattered if I had done so many in such a short space of time. She commented that it would likely skew things somewhat but not overly so. I asked her to explain. She told me that what was being measured was not my interpretation of the images.
She explained that what she was looking for were things like
How quickly I made up my mind
(decisivness/recklessness vs studious/thourogh)
and
Did I Stick to my choice ?
Apparently people will often offer two or three possibles hoping for a response from the tester that will help them justify their final choice. Also a lot of people will observe the testers note taking demenaour and decide that its either negative or positive and decide to change their choice then.
Now because the test was being conducted on behalf of a large company this may not be the normal procedure but it makes sense to be that a test like this can be use to identify rash decision makers and compare them against more reserved (plodding) individuals. Possibly you can guage a testee’s self confidence this way two.
There were a number of other things she mentioned that she was measuring , all in a similar vein , and nothing whatsoever to do with what I percieved in the pictures.
Obviously one of them wasnt memory retention because I cant for the life of me remember what the rest were and I did get the job … I find it far easier to remember her voice and eyes than anything else…ah Ms. Savolainen ! Where are you now ? …sigh